This article presents a systematic quantitative performance analysis for large finite element computations on extreme scale computing systems. Three parallel iterative solvers for the Stokes system, discretized by low order tetrahedral elements, are compared with respect to their numerical efficiency and their scalability running on up to 786 432 parallel threads. A genuine multigrid method for...

For many large-scale applications, solving large sparse linear systems is the most time-consuming part. The important criteria for a suitable solver include efficiency, robustness, and good parallel performance. The Parallel Algebraic Recursive Multilevel Solver (pARMS) [8] is a suite of distributed-memory iterative accelerators and preconditioners targeting the solution of general sparse linea...

Sampling from high-dimensional systems often suffers from the curse of dimensionality. In this paper, we explored the use of sequential structures in sampling from high-dimensional systems with an aim at eliminating the curse of dimensionality, and proposed an algorithm, so-called sequential parallel tempering as an extension of parallel tempering. The algorithm was tested with the witch's hat ...

Evaluation of prosody transformation systems is an important issue. First, the existing evaluation methodologies focus on parallel evaluation of systems and are not applicable to compare parallel and non-parallel systems. Secondly, these methodologies do not guarantee the independence from other features such as the segmental component. In particular, its influence cannot be neglected during ev...

In this paper we discuss the parallel implementation of the auction algorithm for shortest path problems. We show that both the one-sided and the two-sided versions of the algorithm admit asynchronous implementations. We implemented the parallel schemes for the algorithm on a shared memory machine and tested its efficiency under various degrees of synchronization and for different types of prob...

Purpose – The purpose of this paper is to consider the following problem; if the manager of the parallel network systems wants to add new sub-decision making units (sub-DMUs) to each parallel network system, he/she wants to know how much new fuzzy inputs allocate to new sub-DMUs and how much outputs these new sub-DMUs produce such that the efficiency of each parallel network system improve or p...

Evaluation of link prediction methods is a hard task in very large complex networks because of the inhibitive computational cost. By setting a lower bound of the number of common neighbors (CN), we propose a new framework to efficiently and precisely evaluate the performances of CN-based similarity indices in link prediction for very large heterogeneous networks. Specifically, we propose a fast...

This Article is brought to you for free and open access by the Computer Science at ODU Digital Commons. It has been accepted for inclusion in Computer Science Faculty Publications by an authorized administrator of ODU Digital Commons. For more information, please contact [email protected]. Repository Citation Bhagavathi, D.; Gurla, H.; Olariu, S.; Schwing, J. L.; and Zhang, J., "Timeand Co...

Standard Clock (SC) simulation is an efficient form of discreteevent simulation that is applicable to systems with exponential interevent times. It achieves its efficiency by using a common event stream to construct, on sequential or parallel machines, many sample paths in parallel. However, on sequential machines the efficiency is limited by two factors: (1) A fundamental simulation method-rel...